EUVIMEDEUVIMED entdecken
Crossref

A new Multi-Classifier model for complex systems based on belief rule base

Tianhao Zhang, Jiaxing Li, Wenkai Zhou, Xiping Duan, Ning Ma, Yuhe Wang

Engineering Research Express · 01.08.2026 · 8 (15) · 155223

DOI: 10.1088/2631-8695/ae92ea ↗

Kurz gefasst

Fachartikel in der Zeitschrift Engineering Research Express, online erschienen am 10.08.2026. Verfasst von Tianhao Zhang und 5 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Gefördert von: Harbin Normal University Ph.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In complex environments, reliable modeling methods are crucial. belief rule base (BRB) systems, with their excellent uncertainty handling, high transparency, and interpretability, have been widely applied in complex system modeling. However, when the number of input features increases, the rule combinations in BRB grow exponentially, leading to the problem of combinatorial explosion. To address these issues, this paper proposes a Multi-Classifier BRB model (MC-BRB) that aims to maintain prediction accuracy while mitigating the combinatorial explosion in classification tasks. The core idea of MC-BRB is to construct corresponding sub-BRB models based on the number of classes and use SHAP to explain feature importance, thereby performing effective feature selection based on quantitative results. The final prediction result is derived using a soft voting ensemble mechanism. Additionally, to further improve prediction accuracy, this paper introduces an Alpha Evolution (AE) algorithm integrated with hyperplane projection operations (P-AE) to optimize the BRB parameters. Through four benchmark experiments and a practical gearbox fault diagnosis case study, the results show that the proposed MC-BRB model effectively mitigates the combinatorial explosion issue while maintaining high accuracy and enhancing the model’s stability.

Förderung

  • Harbin Normal University Ph

Zitieren

APA

Zhang, T., Li, J., Zhou, W., Duan, X., Ma, N., & Wang, Y. (2026). A new Multi-Classifier model for complex systems based on belief rule base. Engineering Research Express, 8(15), 155223. https://doi.org/10.1088/2631-8695/ae92ea

Vancouver

Zhang T, Li J, Zhou W, Duan X, Ma N, Wang Y. A new Multi-Classifier model for complex systems based on belief rule base. Engineering Research Express. 2026;8(15):155223. doi:10.1088/2631-8695/ae92ea

BibTeX
@article{Zhang2026,
  author = {Zhang, Tianhao and Li, Jiaxing and Zhou, Wenkai and Duan, Xiping and Ma, Ning and Wang, Yuhe},
  title = {A new Multi-Classifier model for complex systems based on belief rule base},
  journal = {Engineering Research Express},
  year = {2026},
  volume = {8},
  number = {15},
  pages = {155223},
  doi = {10.1088/2631-8695/ae92ea}
}
Kostenlos starten →